[aphelion]

Hi. Will this destroy your career to turn this into an actual book?

[Constitution45]

Haha this is too much red pill, I think I need to watch a few seasons of Sex and the City to restore my balance.

[Robert Hedrock]

China shocks the world by genetically engineering human embryos

http://www.telegraph.co.uk/news/science/...bryos.html

(07-30-2014 05:30 AM)Sp5 Wrote:  

Globalism will bring a new kind of nexus between culture, commerce and technology. It's not just the technology, it's how and who uses it. Nemencine is right to focus on the likely Chinese use of these new human-enhancing discoveries. The Chinese seem to pick their rulers through a rigorous screening process.

[Kabal]

Unfortunately, I either wasn't reading the forum or didn't read this thread when this was originally posted.

I figured I would start with #4 on IQ.

I ended up deciding to stop with #4 as well, because I immediately spotted something wrong.

(07-27-2014 03:28 PM)Nemencine Wrote:  

#4. ON IQ SCORES.

There have been a lot of electronic ink spilled over IQ ratios and patterns, asian IQs, black IQs, white IQs, hispanic IQs, etcetera. 2Wycked did a solid review post on GATTACA, and the discussion on that thread turned towards the genetic basis of intelligence. Speaking of genetic intelligence, here is a very disruptive news for society:

There is a gene present in many different mammals(including humans) that plays a solid role in genetic intelligence and memory. That gene is NR2B. NR2B stands for Glutamate [NMDA] receptor subunit epsilon-2 gene, or N-methyl D-aspartate receptor subtype 2B.

When scientists took this gene/protein and over-express it, that is, they modified the gene so that the body produce more of this protein(NR2B) above normal levels. What happened? The test subject, affectionately named Hobbie-J, started to exhibit "improved memory performances in novel object recognition test, spatial water maze, and delayed-to-nonmatch working memory test." Basically, it becomes 3 times smarter than average.

For the sake of comparison, if we say that the average IQ of rat is 100; Hobbie-J has an IQ of 300. Guess what? Humans have that exact same gene doing the exact same thing in us.

I have taken the liberty of finding a layman's explanation of what scientists did, you can find it here. For the RVF science majors, here is the detailed scientific publication.

You just took at face value the journalist’s claim that the transgenic operation increases rat memory length by three times, took that number 3, and applied it to initial IQ to arrive at implied post-operation IQ?

Even accepting the journalist’s claim, why would you ever think this is the proper calculation?



The IQ distribution of ~N(100, 15^2) (mean of 100, variance 15^2 [standard deviation 15]) is just a location and scale adjustment to the standard normal. A 100 to 300 increase in IQ is identical to saying a 13 and 1/3 standard deviation increase ((300 – 100) / 15 = 13 and 1/3). What part of a threefold (3x) increase in memory length would lead you to conclude a 13 and 1/3 standard deviation increase in IQ, even if we accepted memory length as the perfect metric for IQ?

Your attempt at calculating an implied IQ increase from the transgenic operation looks like something a desperate undergraduate did as a hailmary while his time was running out on his Introduction to Psychology exam.

You took the journalist’s remark that memory length increased threefold to mean a threefold increase in “smartness,” which is an unfounded translation. However, even reporting a threefold increase in memory length is misleading.

The threefold outperformance of transgenic vs. wild-type rats were specific to the time gaps of 1 and 3 days; that is, the scalar that links the ratio between the transgenic rats' memory lengths to wild-type rats' memory lengths is dependent on the time-frame we are looking at and the test. It (the 3) does/would not generalize to arbitrary time lengths, e.g. 3 and 9 days. At 7 days transgenic rats did not outperform wild-types at all. The journalist should have been explicit on that.

The original study (Wang. et al.) observed that the transgenic rats discriminated between "novel" and non-novel objects at a higher rate (as inferred from "exploratory preference") after three days than wild-type rats did after one-day; that is, we say that the transgenic rats had better memory recollection after three days than wild-types had after one, using novel object discrimination as the measurement.

A more complete (and accurate) description of the finding would be, eye-balling their graph: After one hour, transgenic rats discriminated in favor of the novel object at 70%, compared to about 66% for wild-type rats--the difference was statistically insignificant. After one day, transgenic rats discriminated in favor of the novel object at 65%, compared to about 53% for wild-type rats--difference significant. After three days, this was 62%, and non-significantly lower than 50%, respectively--difference significant. After 7 days, both, were non-significantly different than 50%.

Something similar is found for 1 and 3 minutes on a different test, but is also subject to the (non) generalized time scale issue. And the fact that we're looking at samples of mice in the paper, not "Hobbie-J."

Not nearly as exciting as reporting a generalized threefold increase in memory length, I suppose.

This is sub-optimal handling by the journalist in terms of precision, but as a scientist, you should know better than to take scientific journalism at face value—especially given the constant mini-brouhahas over miscommunication in scientific journalism in recent years.

Your chain of "logic" went like this:

1. The study found that transgenic mice had memories after three days comparable to or better than wild-type mice after one day on one test (and three minutes vs. one on another test).
2. The operation increases rat-memory lengths three-fold.
3. The operation makes rats 3 times smarter.
4. The operation increases arithmetically rat IQ three-fold, IQ being ~N(100, 15^2).
5. The operation would do the same in humans.

The jump from 1 to 2 is misleading and unclear, the journalist could have been more precise. One can live with that, though.

The jumps from 2 to 4 are errors on your part.

The jump from 4 to 5 is completlely unfounded—the gene does the exact same thing in humans, as in qualitatively and quantitatively identical?

Genome studies have concluded that human IQ is like height—numerous variants with small effect sizes. A gene purported to make a large increase in IQ, something like 2 (as in 100 to 102), would need a ton of justification and replication, much less one purported to make an increase of 200 (from 100 to 300).

You didn't even understand properly a source you described as fitting for "laymen," and it does not appear that you read the original paper. And it's evident you don't understand IQ as well as you should.

To position yourself as a brave renegade combating the ignorance of "The Manosphere"--and to claim that you are limiting your discussion to scientific facts and findings--and then immediately bungling the science, is more befitting a clown than a scientist.

Another Don Quixote fighting windmills.

If this is indicative of the rest of your series in this thread, it means it is riddled with elementary errors.

If you work professionally as a scientist and you are not-infrequently this sloppy, you should consider if your colleagues are too nice to tell you that you have no clothes. I hope for your sake, not, though.

I have remarked multiple times to forum members in person that posts on RVF are frequently just illustrations of the Dunning-Kruger Effect. Much confidence with little competence.

It appears that I have another thread/post to add as a data point.

I normally wouldn’t be this harsh (most of the time, I don’t bother responding to mathematical/statistical/scientific errors on this board, just too time consuming), but this one was particularly egregious since you present yourself as a subject matter expert to an audience that gives you the benefit of the doubt.

[Huey]